Performance and emissions of a DISI engine fueled with gasoline/ethanol and gasoline/C-4 oxygenate blends - Development of a PM index correlation for particulate matter emission assessment

The increasing attention devoted to biomass-derived oxygenated molecules is among others related to their potential to reduce pollutant emissions at the exhaust of spark-ignition engines particularly when combined with the latest direct injection technologies. The impact of adding such compounds to gasoline on engine performances, however, strongly depends on the composition, the chemical structure and the physical properties of the considered additives. The formation of particulate matter (PM), whose emissions are subject to increasingly stringent standards, is especially influenced by such factors thus prompting the need for the formulation of predictive multi-parameter PM indexes helping to support the development of advanced and cleaner engine fuels. Within this context, the present work proposes to analyze the performance and emissions of a direct injection spark-ignition engine fueled with mixtures made of gasoline with five different oxygenated fuels under homogeneous injection condition. Ethanol and four C-4 oxygenates (1-butanol, butanone, butanal and methyl propionate) have been selected to be blended with gasoline in proportions of 10, 20 and 40 vol%. Engine performances and exhaust emissions have been characterized in terms of specific fuel consumption, flame development angle, fully developed combustion duration and concentrations of CO, HC, NOx and PM. While the addition of oxygenated fuels increases the specific fuel consumption proportionally to the mixture lower heating value, it tends to decrease the flame development angle while not significantly influencing the fully developed combustion phase. Besides, relatively unchanged CO and HC emissions have been observed contrary to NOx and PM releases which have been found to decline with increasing oxygenate contents. Measured particulate concentrations have been compared with the predictions of four PM and sooting indexes from the literature including the Particulate Matter Index (PMI), the Threshold Sooting Index (TSI), the Oxygen Extended Sooting Index (OESI) and the Fuel Equivalent Sooting Index (FESI). While the FESI has been found to be the most efficient to reproduce experimentally monitored trends, it has still been coupled with five different fuel properties and volatility metrics to propose, through a multi-parameter optimization procedure, a generic PM index correlation allowing reproducing well measured PM emissions. By this way, it has been shown that the heat of vaporization and the fuel energy content are key factors to be combined with a sooting propensity indicator so as to capture engine exhaust PM releases.